Home

HILDA User Manual
Overview The HILDA Data Users How to Order the Data Security Requirements How the Data Files are Provided Structure of the Data Files Identifiers Matching Files Creating Longitudinal Files Variable Name Conventions Missing Value Conventions Data with Negative Values Confidentialisation Derived Variables Age and Sex History Spatial Current Education Current Marriage /Defacto Relationship Children Child Care Employment Calculating hourly wage rates Calendar Family Health Housing Time Use Household Expenditure Life – Personality Income and Income Imputation Wealth and Wealth Imputation Weights Documentation Data Quality Issues The HILDA Sample Data Collection Pilot Testing Questionnaire Length Interviewers Fieldwork Process Response Rates Attrition Bias Data Processing HILDA User Training Getting More Information References Appendices Summary of HILDA Survey content, waves 1 to 6 HILDA Survey Instrument Development and Sources Changes made to the data between releases Little and Su method About the HILDA User Manual HILDA Survey Home

Missing Value Conventions

Global codes are used throughout the dataset to identify missing data. These codes are not restated for each variable in the coding framework.

Numeric Variables

All missing numeric data are coded into the following set of negative values.

Table 12: Missing value conventions for numeric variables

*Code*	*Description*
–1	Not asked: question skipped due to answer to a preceding question
–2	Not applicable
–3	Don’t know
–4	Refused or not answered
–5	Invalid multiple response (SCQ only)
–6	Value implausible (as determined after intensive checking)
–7	Unable to determine value
–8	No Self–Completion Questionnaire returned and matched to individual record
–9	Non–responding household
–10	Non–responding person (Combined File only)

Note that the SPSS files have these global missing values (–10 to –1) set to SPSS user–defined missing. To turn off this setting for an individual variable use "MISSING VALUES varname1 ()." To turn off this setting for all variables (for example, if you need to include those who are coded as –1 ‘Not asked’) use the following code:

set errors=none.
do repeat x=all.
missing values x ().
edn repeat.
set errors=listing.
execute.

Text Variables

Text variables with missing values will typically contain the following text as shown in the table below.

Table 13: Missing value convention for text variables

*Text*	*Description*
[blank]	Missing information (no reason specified)
–1	Not asked
–2	Not applicable
–3	Don’t know
–4	Refused
–7	Unable to determine value
–9	Non–responding household
99	Not given (new person)

top of page